Study of anisotropy in nuclear magnetic resonance relaxation times of water protons in skeletal muscle

Study of anisotropy in nuclear magnetic resonance relaxation times of water protons in skeletal muscle

STUDY OF ANISOTROPY IN NUCLEAR MAGNETIC RESONANCE RELAXATION TIMES OF WATER PROTONS IN SKELETAL MUSCLE S. R. KASTURI, Tata Institute of Fundamental Re...

868KB Sizes 0 Downloads 97 Views

STUDY OF ANISOTROPY IN NUCLEAR MAGNETIC RESONANCE RELAXATION TIMES OF WATER PROTONS IN SKELETAL MUSCLE S. R. KASTURI, Tata Institute of Fundamental Research, Colaba, Bombay, 400005, India D. C. CHANG AND C. F. HAZLEWOOD, Departments ofPediatrics and Physiology, Baylor College of Medicine, Houston, Texas 77030, and Department of Physics, Rice University, Houston, Texas 77005 U.S.A. ABSTRACr The anisotropy of the spin-lattice relaxation time (T,) and the spin-spin relaxation times (T2) of water protons in skeletal muscle tissue have been studied by the spin-echo technique. Both T, and T2 have been measured for the water protons of the tibialis anterior muscle of mature male rats for 0 = 0, 55, and 900, where 0 is the orientation of the muscle fiber with respect to the static field. The anisotropy in T, and T2 has been measured at temperatures of 28, - 5 and - I0°C. No significant anisotropy was observed in the T, of the tissue water, while an average anisotropy of -5% was observed in T2 at room temperature. The average anisotropy of T2 at -5 and -1 0°C was found to be -2 and 1.3%, respectively.

INTRODUCTION

Water and ions are major constituents of a living cell. In recent years, there has been considerable interest in the study of the physico-chemical state of these constituents using nuclear magnetic resonance (NMR) (1-28) and other physico-chemical techniques (1,29, 30-43). The concept of "ordering" of water has been suggested in some of these studies (4,10,11,15,24,42), and the hypothesis has been advanced that interactions between the water molecules and the biopolymer interface result in a reduction in the mobility of the water molecules in the cell. This hypothesis implies that the correlation times (rotational and/or translational) for these water molecules are longer in tissues than those of bulk water. NMR experiments show that partial orientation (or ordering) of the water molecules due to these interactions exists in hydrated collagen (44-48), oriented DNA (49), oriented rayon fibers (50), frog muscle tissue (51), and other systems (52). In nerve tissue, however, there is some controversy regarding the existence of oriented water (52,53). Testing for the presence of oriented water, Finch et al. (9) studied the water in resting muscle and observed no anisotropic effect on either T2 or the diffusion coefficient of cellular water.' In contrast, Cleveland et al. (24) observed a significant anisotropy (28%) in the 'In a personal communication to Dr. Haziewood, Finch indicated that his attempts to measure anisotropy were hampered by the fact that he could not be certain that he could properly align small pieces of muscle in the NMR sample tubes. BIOPHYS. J. © Biophysical Society Volume 30 June 1980 369-382

*

0006-3495/80/06/369/13 $1.00 369

diffusion coefficients of intracellular water in skeletal muscle measured with the muscle f'iber orientation parallel and perpendicular to the direction of diffusion. In our earlier studies, we observed an anisotropy in the spin-spin relaxation time of water in rat skeletal muscle.2 Fung also observed a similar anisotropy in the T2 water in frog gastrocnemius muscle (51). We suspect that the observed anisotropy in the relaxation time results from the interaction between water molecules and biomacromolecules (especially the contractile proteins). It would seem that the anisotropic effect should be enhanced in a system in which the water-macromolecular interactions are greatest. It is known that cellular water enters a supercooled phase when the tissue is cooled below OOC.3 At lower temperatures, when most of the tissue water is frozen, there is still a fraction of water that does not freeze. This nonfrozen water is generally thought to be hydration water, and the changes in freezing properties are attributed to the water-macromolecular interaction. The nonfrozen water thus offers a convenient system for studying the detailed mechanisms of the interaction between the water molecules and the protein surface. In hopes of gaining further insight into the nature of the anisotropy in the relaxation times and its implications for the ordering of water molecules in muscle, we have carried out an investigation of this anisotropy at three different temperatures (28, -5, and - IOC). METHODS

NMR Measurements The spin-lattice relaxation time (T,) and spin-spin relaxation time (T2) of the intracellular water protons were measured using a Bruker SXP spectrometer (Bruker Instruments, Inc., Billerica, Mass.) equipped with a frequency synthesizer and a 12-in. electromagnet which has a 1.75-in. air gap. Measurements of T, and T2, were made at room temperature, at - 50C, and at -1 OOC using a homemade temperature controller. The spectrometer was operated at a frequency of 30.3 MHz in this study. The T1 was measured using the 1800-r-900 pulse sequence (54) and by monitoring the full recovery of the magnetization to equilibrium value. The T2 was measured using the Carr-Purcell-Meiboom-Gill (CPMG) method (55). The 2r spacing was chosen between 6 ms (at room temperature) and 0.8 ms (at T = -10°C). All data of relaxation measurement, both T1 and T2, were analyzed using a program based on an iterative curve fitting technique (15). The results reported in this paper are the initial relaxation times which are determined from the initial decay of the magnetization signal. ("Initial decay" means the tangential line of the spin-echo decay curve at t = 0. Usually the first 20% of the echo decay curve is linear and can be used for the initial decay fit). The linearity of the detection system was carefully compensated, and all data were normalized by computer against a standard calibration curve. The temperature of the sample could be measured and maintained to an accuracy of 0.50C. All T2 measurements were signal-averaged for 32 repetitive scans. At room temperature, the signal-to-noise ratio was better than 2,000 to 1. At subzero temperature, the signal-to-noise ratio was reduced to -500 to 1 (due to freezing). The number of repetitive scans in T, measurements was eight. The signal-to-noise ratio was similar to that of the T2 measurements, within a factor of 2. Although our magnetic field was not locked, it drifted less than 100 Hz in 30 min. Since we used diode-detection in all measurements and our radio frequency pulses had an equivalent bandwidth of 100 kHz, this small field drift did not result in any observable instability in the measured signal.

Sample Preparation The skeletal muscles used in this study were obtained from white rats (Texas Inbred Mouse Company, Houston, Tex.) weighing between 260 and 530 g. The rats were killed by cervical fracture, the tibialis 2Chang, D. C., C. F. Hazlewood, and D. E. Woessner. Unpublished data. 3Fogal, J. M., D. C. Chang, C. F. Hazlewood, and H. E. Rorschach. Manuscript to be submitted for publication. 370

BIOPHYSICAL JOURNAL VOLUME 30 1980

anterior muscle was quickly excised, and the gross connective tissue was removed from the tissue. The muscle orientation (0), which is the angle of the fibers, with reference to the direction of the static magnetic field, was managed as follows (24): a piece of 6-mm (i.d.) glass tubing was attached perpendicular to a glass rod, and this system was placed inside a 10-mm o.d. NMR glass tube so that the muscle sample, held in the 6-mm glass tubing, fitted flush inside the 10-mm o.d. glass tube. The excised muscle was drawn through the 6-mm tube until a linear portion of the muscle was held in place, and the excess portion of the muscle was then trimmed off at both ends. This sample was then placed in the 10-mm o.d. NMR glass tube mentioned above. The orientation of the muscle fiber could be noted at the top of the 10-mm glass tube and was varied by manually rotating the 10-mm glass tube. The measurements were started within 5 to 15 min after the sample was excised. Some samples were stored overnight in the refrigerator; no effect of this procedure on anisotropy was observed.

Sample Cooling Each muscle preparation was cooled by blowing cold nitrogen gas onto the sample. Most of the samples were cooled by cooling the sample from 0 to -200C within 30 to 45 min. Some samples (as marked in the results) were cooled at slower rates. For these samples, a minimum of 2 h was required to reach the desired temperature between 0 and - 200C. To insure that most of the tissue was frozen, the sample was first cooled to -200C (nucleation occurs at T=- - 140C) and then brought back to the desired TABLE I

ANGULAR DEPENDENCE OF THE T2 OF CELLULAR WATER PROTONS AT 280C IN SKELETAL MUSCLE. ALSO SHOWN IS THE ANGULAR DEPENDENCE OF T2 IN AGAR GEL SAMPLE UNDER IDENTICAL CONDITIONS

A4

Sample No. 0

-



(ms) 1* 2 3 4 5 6 7 8 9 10

11 12 13 14

15* 16 17 18 19 20 21 22 Agar gel

39.3 44.6 51.8 47.0 49.3 46.4

0.3 0.5 0.8 0.4 0.8 0.6 0.6 0.3 45.9 + 0.7 42.1 ± 0.4 40.8 + 0.4 39.6 ± 0.3 42.0 ± 0.6 41.9 ± 0.4 48.9 ± 0.6 40.2 ± 0.4 43.3 ± 0.5 41.0 ± 0.4 42.0 ± 0.4 42.6 ± 0.4 41.6 ± 0.5 P << 0.001 30.1 ± 0.1 ± ± ± ± ± ± 45.5 ± 41.3 ±

0

=

550

0 = 900

(ms)

(ms)

43.4 ± 0.5 45.1 ± 0.6 54.4 ± 0.9 48.6 ± 0.6 50.2 ± 0.9 47.7 ± 0.7 47.0 ± 0.6 42.8 ± 0.3 46.7 ± 0.6 43.7 ± 0.4 44.4 ± 0.5 42.7 ± 0.3 43.2 ± 0.5 44.0 ± 0.5 50.6 ± 0.7 44.3 ± 0.5 45.6 ± 0.5 44.4 ± 0.4 45.3 ± 0.4 44.0 ± 0.3 43.4 ± 0.5 Mean ± SE 30.0 ± 0.08

40.1 ± 0.3 44.9 ± 0.6 53.4 ± 0.7

46.9 50.4 47.8 46.3 42.7 45.7 42.9 44.4 43.3 42.5 44.0 51.1 44.3 44.9 43.0 45.1 44.0 42.8

± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±

0.5 0.8 0.7 0.5 0.3 0.5 0.4 0.4 0.4 0.5 0.4 0.7 0.4 0.4 0.3 0.4 0.3 0.3

29.9 ± 0.1

1.104 ± 0.015 1.011 ± 0;018 1.050 ± 0.024 1.034 ± 0.015 1.018 ± 0.025 1.028 ± 0.020 1.033 ± 0.019 1.036 + 0.010 1.017 ± 0.020 1.038 ± 0.013 1.088 + 0.015 1.078 ± 0.010 1.029 + 0.019 1.050 ± 0.016 1.035 ± 0.019 1.102 ± 0.017 1.053 ± 0.017 1.083 ± 0.014 1.079 ± 0.014 1.033 ± 0.012 1.043 ± 0.017 1.050 ± 0.006 0.997 ± 0.004

*Sample I is EDL muscle and sample no. 15 is soleus muscle.

tA2 is defined as A2 = T2 (0 = 550)/T2 (0 = 00) (See Text). KASTURI ET AL. Anisotropy in NMR Relaxation Times

371

temperature. No significant difference was observed between the results obtained using the rapid and the slow cooling methods.

RESULTS

The results of the measurement at room temperature (280C) of T2 of tissue water protons in tibialis anterior muscle tissue for 6 = 0, 55, and 900 are shown in Table I. Table I also contains one study on the extensor digitorum longus (EDL) muscle and one on the soleus muscle. The results for a model system (3% agar gel sample) are also given in Table I. The measurements for the agar gel sample were conducted under identical conditions using the same sample holder and same size sample, etc. In the last column of this table are the values for the anisotropy parameter, A2, which we define as follows:

A- T2(550) T2(00)

(1)

where T2 (550) and T2 (00) are in the spin-spin relaxation times for the muscle orientations 0 = 550 and 0 = 00, respectively. An anisotropy parameter, A1, for the spin-lattice relaxation time, T,, is similarly defined. The anisotropy parameter is defined as above since one would expect T2 to be longest for 0 = 550 and shortest for 6 = 00, as will be discussed later. The last column of Table I shows a definite anisotropy in every muscle sample investigated. The average value of A2 is 1.050 ± 0.006, which represents an anisotropy of 5%. These data should be compared

10.0 I-

..A 0,

z

h'

0-55

1.0 I-

Q.5 cL

to acJ

cLI a;GO

0

Q.,

Lu-

0l01

I

I m.

....

0'/3O

I

I

.iiI

I

I

I

I

ii.

1.

60 -9S1 1-50 1S021O240 TI TME :s)--:-... *.

...........

FIGURE I Echo decay of water protons using the Carr-Purcell-Meiboom-Gill (GPMG) pulse sequence. 0 is the angle between the sample axis and the external magnetic field. (A) Muscle sample at 0 = 550. (B) Muscle sample at 0 = 00. (C) Agar gel sample under identical conditions at 0 = 00 and 0 = 550. All the data shown were measured at T = 280C.

372

BIOPHYSICAL JOURNAL VOLUME 30

1980

with the datum on the 3% agar gel system in which no anisotropy was observed. A typical proton spin-echo decay at 280C is shown in Fig. 1. This muscle tissue was measured using a CPMG sequence for the two orientations of 55 and 00. Also shown in the same figure is the spin-echo decay curve for the water protons in the 3% agar gel sample for the same two orientations. This figure clearly shows that there is no anisotropy in the agar-water model system, while there is an anisotropy in the muscle tissue. The data in Table I were subjected to a statistical test to determine the probability, P, that the mean difference between the T2 (550) and T2 (00) in the paired observations occurs by chance. For the data in Table I, the probability, P, is much less than 0.001. Table II shows the T2 values for muscle protons for different orientations of the muscle fiber at a temperature (7) of - 5°C; Table III gives similar results at T = -10°C. At T = - 50C there is an average anisotropy of 2%, while at T = -10°C, the average anisotropy is 1.3%. A statistical analysis of the data shows that the mean difference between the values of T2 (550) and T2 (00) in paired observations at T = -5SC is statistically significant (P << 0.001), while at T = - 10°C, the statistical significance is 0.05 > P > 0.025. The larger P value observed in the latter case is probably due to the poor signal-to-noise ratio at -10°C. The typical spin-echo decay curves for water protons in the muscle tissue for orientations 0 = 550 and 0 = 00 at -5 and - 10°C are presented in Figs. 2 and 3. A small but definite anisotropy in the T2 is apparent at both temperatures. In the control agar gel sample, no echo signal was observed at these temperatures because the amount of the nonfrozen water in the control sample is extremely small. -

TABLE II

ANGULAR DEPENDENCE OF T2 OF CELLULAR WATER PROTONS IN SKELETAL MUSCLE AT T = -50C T2

Sample No. 0 = 00

1 2 3 4 5 6 7 8 9 10 11 12 13 14* 15 16*

17*

(ms) + 0.04

5.85 5.40 5.76 6.02 5.15 5.53 5.36 5.67 5.82 5.35 5.77 5.16 6.18 5.81 6.03 5.29 5.50

± ± ± ±

±

+ ±

± ± ±

± ± ± ±

+ ±

0.05 0.06 0.10 0.02 0.04 0.07 0.05 0.05 0.08 0.03 0.05 0.04 0.04 0.04 0.04 0.04

0 = 550

A2 0 = 900

(ms)

(ms)

5.92 ± 0.05 5.37 + 0.03 5.83 ± 0.06 5.98 ± 0.07 5.24 ± 0.03 5.82 ± 0.05 5.50 + 0.08 5.83 ± 0.05 5.92 ± 0.04 5.27 ± 0.04 5.81 ± 0.03 5.33 ± 0.05 6.30 ± 0.03 6.08 ± 0.03 6.22 ± 0.04 5.57 ± 0.04 5.71 ± 0.05 P << 0.01

5.94 ± 0.04 5.16 ± 0.05 5.65 ± 0.05 5.84 ± 0.07 5.17 ± 0.04 5.66 ± 0.04 5.50 + 0.09 5.71 ± 0.05 5.78 ± 0.05 5.17 ± 0.06 5.59 ± 0.04 5.24 ± 0.05 6.19 ± 0.04 6.01 ± 0.04 5.98 ± 0.04 5.51 ± 0.04 5.80 ± 0.05 Mean ± SE

1.012 ± 0.994 ± 1.012 ± 0.993 ± 1.017 ± 1.052 ± 1.026 + 1.028 ± 1.017 ± 0.985 ± 1.007 ± 1.033 ± 1.019 ± 1.047 ± 1.032 ± 1.053 ± 1.038 ± 1.021 ±

0.011 0.011 0.015 0.020 0.007 0.012 0.020 0.013 0.011 0.017 0.007 0.014 0.008 0.009 0.010 0.011 0.012 0.005

*Samples 14, 16, and 17 have been cooled slowly at the rate mentioned in the text.

KASTURI ET AL. Anisotropy in NMR Relaxation Times

373

.

TABLE III ANGULAR DEPENDENCE OF T2 OF CELLULAR WATER PROTONS IN SKELETAL MUSCLE AT T= -100C

Sample No. 0

=

00

0

(ms) 1 2 3 4 5 6 7 8 9 10 11* 12

13* 14*

=

550

(ms)

3.06 + 0.03 3.42 + 0.08 3.21 + 0.07 3.27 + 0.07 3.02 ± 0.04 3.49 + 0.10 2.82 + 0.04 3.17 ± 0.05 2.85 + 0.06 3.20 + 0.04 3.26 + 0.04 3.16 ± 0.05 3.04 + 0.04 3.06 + 0.04 0.05 > P> 0.025

3.27 + 3.58 + 3.16 ± 3.34 + 3.03 ± 3.56 ± 2.87 ± 3.13 + 2.80 ± 3.22 + 3.33 + 3.22 + 3.07 ± 3.02 +

A2

0 = 900

(ms)

3.05 + 0.04 3.36 ± 0.06 3.13 ± 0.05 3.25 + 0.05 2.99 ± 0.04 3.28 + 0.07 2.80 ± 0.04 3.24 + 0.06 2.74 ± 0.05 3.19 + 0.04 3.29 + 0.04 3.21 + 0.04 3.06 ± 0.04 3.00 ± 0.05 Mean + SE

0.04 0.09 0.04 0.04 0.04 0.09 0.06 0.05 0.04 0.03 0.03 0.03 0.02 0.07

1.069 1.047 0.984 1.021 1.003 1.020 1.018 0.987 0.982 1.006 1.022 1.019 1.010 0.987 1.013

0.17 + 0.076 ± 0.025 + 0.029 + 0.019 + 0.039 ± 0.026 ± 0.022 ± 0.025 + 0.016 ± 0.016 ± 0.019 + 0.015 ± 0.026 + 0.007 ±

*SampIeS 11, 13, and 14 are cooled slowly at the rate mentioned in the text.

The results of the measurement of the proton T, for muscle orientations 6 = 0, 55, and 900 at T = 280C are shown in Table IV. The anisotropy parameter, Al, is defined in a way similar to A2. The results in this table show that the anisotropy is very small (0.7%). The mean difference of the T, values at 6 = 00 and 0 = 550 is, however, statistically significant. The WI)

W-1 z

:T -"

5.0

.C:

...

'..

:.,

4.-A

ca. 1--

IC

50. I.C

C-

:4c C.)

'0.1. 0.0

a

550

70|.;;

. §.

*-

_

o:k

<\

F

.' ... wS :.:

4

M8 12 TIME

FIGURE 2 Echo decay of cellular water protons -50C for 0 = 0° and = 550

374

-.e..0m _

in

\:

24

16

26

(ms)--*.

skeletal

muscle

using the CPMG

sequence at

T=

BIOPHYSICAL JOURNAL VOLUME 30

1980

Z. 2.0~~~~~~~~~~~~~~~0 -1

.

I-

v

z 1.0 -.

.

....

.;.

iF * S . l. .. ... S >50

'';_'@1'.0

.'.'

''

~~~~~-'8 .s55°

= : \

Lu5

_

~0j

.,

0. 0

:

*.

:. i

0 I 0~~~4 z~~~~~~

FIGURE 3 Echo decay of cellular water protons in skeletal muscle usi ng the CPMG sequence at T= - IO0Cfor=0and= 550r

angular dependence of the T, data at -5 and - 0OC iS Summarized in Table V. At T-Q50C, the mean difference between T, (550) and T, (00) is not statistically significant (PF> 05); the results are similar for T = - 100C (0.5 >Pu>0.25). The average anisotropy at these two temperatures is 0.2% and is within experimental error. TABLE IV

ANGULAR DEPENDENCE OF T, OF CELLULAR WATER PROTONS IN SKELETAL MUSCLE AT 280C

Sample No.

1 2 3 4 5 6 7 8 9 10

*A is defined as A,

A, 0 =00

0 = 550

(ms) 668 + 2 636 ± 1 705 + 1 704 + 1 685+1 675 + 1 658 + 1 652 + 1 662 ± 1 668 ± 1 0.005 > P > 0.001

(ms)

(ms)

673 + 5 636 ± 2 713 + 1 704 + 1 693+1 677 + 1 666 + 1 652 ± 1 670 + 1 675 ± 1

674 ± 3 636 + 1 713 ± 1 705 ± 1 691±1 678 ± 1 665 ± 1 651 +±1 674 + 1 674 +±1 Mean ± SE

= T, (O

=

0

900

1.008 1.000 1.011 1.000

± 0.008 ± 0.004 ± 0.002

± 0.002

1.012±0.002 1.003 1.012 1.000 1.012 1.010 1.007

± 0.002 + 0.002 ± 0.002 ± 0.002 ± 0.002 ± 0.002

550)/T, (0 - 00).

KASTURI ET AL. Anisotropy in NMR Relaxation Times

375

TABLE V ANGULAR DEPENDENCE OF T, OF CELLULAR WATER PROTONS IN SKELETAL MUSCLE AT -5 and -100C

Sample No.

Temperature

1 2 3 4 5 6 7 8

-50C - 50C

(T)

0=00

(ms)

1 2 3 4 5 6

-50C -50C -50C -50C -50C -50C

-10C -10°C

-10°C -10°C -10°C -10°C

120.0 122.2 114.2 115.5 120.9 118.5 113.4 120.9

78.7 74.8 78.3 75.8 72.8 76.0

± ± ± ± ± ± ± ±

T,A

0=550

(ms) 0.5 0.6 1.6 0.7 0.7 1.0 1.3 0.8

± 0.6 ± 1.2 ± 0.8 ± 0.8 ± 1.9 ± 1.3

0 =900

(ms)

122.0 ± 0.6 122.5 ± 0.7 113.9 ± 1.3 114.9 ± 0.9 119.1 ± 1.1 119.4 + 0.6 113.1 + 1.7 121.0 ± 0.7 P> 0.5

122.0 123.9 114.6 114.9 118.8 120.3 114.6 121.8 Mean

+ 0.6 ± 0.4 ± 1.4 + 1.1 + 1.0 + 1.0

79.0 ± 0.8 75.2 + 0.8 78.4 ± 0.9 75.8 + 1.0 73.3 + 1.5 75.6 ± 1.3 0.5 > P > 0.25

79.3 74.9 77.9 76.9 73.8 76.1 Mean

+ SE

1.017 1.002 0.997 0.995 0.985 1.008 0.997 1.001 1.002

+ 0.006 0.008 0.018 + 0.010 ± 0.011 ± 0.010 + 0.019 ± 0.009 ± 0.003

+ 0.7 0.9 + 1.0 ± 0.7 ± 1.5 ± 1.2 ± SE

1.004 1.005 1.001 1.000 1.007 0.995 1.002

0.013 + 0.019 + 0.015 ± 0.017 + 0.033 ± 0.024 + 0.0018

± 1.6 ± 0.7

±

± ±

±

DISCUSSION

Although a very small anisotropy in the T1 tissue water protons in tibialis anterior muscle tissue was observed at room temperature, none was observed at the subzero temperatures. A definite anisotropy in the T2 of water protons in the muscle tissue, however, was observed. The origin of the anisotropy has been thought to be due to the partial orientation (ordering) of the water molecules as they interact with the actin-myosin filaments (44, 47-51). In broad line NMR studies of hydrated collagen (44, 46) and oriented DNA (49), the angular dependence of the dipolar splitting was found to obey the angular relationship AB = K(3 COS2O

1),

(2)

where AB is the separation of the resonance peaks, K is a constant which depends on sample hydration, and 0 is the angle between the sample fiber axis and the direction of the static field. Since the effective T2 is in general inversely proportional to the dipolar splitting, one may expect that the average T2 will be longest for 9 = 550 and shortest for 0 = 00, with the T2 value for 0 = 900 lying in between. In tibialis anterior muscle fiber, this is precisely what was observed for T2 at 280C. At -5 and -100C, a similar angular dependence was found, although it was not as well defined as that at 280C. These observations suggest the possible existence of water chains along the direction of the muscle fiber (44) or, alternatively, the existence of an anisotropic reorientation of H20 mulecules (46, 50). The preferred orientation of water molecules may or may not be localized. The bulk of the myoplasmic water may be "ordered" to some extent, or, only a small part of the water bound to the surface of the biomacromolecules may be preferentially oriented, but due to an exchange between the "bound" water and the bulk water in the tissue, a new anisotropy of the 376

BIOPHYSICAL JOURNAL VOLUME 30

1980

TABLE VI

TEMPERATURE DEPENDENCE OF ISOTROPIC AND ANISOTROPIC CONTRIBUTIONS TO TRANSVERSE RELAXATION RATE Temperature

tT (55°))av

A2 - I

T2*ni,c

(°C)

(s ') 21.9 174.5 315.4

0.050 ± 0.006 0.021 ± 0.005 0.013 ± 0.007

1.1 ± 0.1

28 -5 -10

P/T2..nisco

(s-') 3.7 ± 0.9 4.1 ± 2.2

1.1 0.93 0.82

*P is normalized to 1.0 at 28°C.

average T2 is observed. If the latter possibility were true for muscle water, then one would expect that a larger anisotropy of T2 at temperatures below 0°C would have been observed, and the anisotropy would have increased as the temperature was lowered, because the isotropically rotating bulk water molecules have the least interaction with muscle proteins and would have frozen at subzero temperatures. We found, however, that the anisotropy decreased as the temperature was lowered. This observation may be explained in two ways; (a) the preferred orientation is not localized in the nonfrozen region, or (b) the isotropic relaxation rate increases faster than the anisotropic relaxation rate as the temperature is decreased. These two interpretations ar0ot mutually exclusive, and we found evidence that both of them are operative to a certain extent. The T2 anisotropy, for example, can be analyzed by noting that the T2 depends both on isotropic and anisotropic terms:

1+ 1(3)

1=

T2

T2,iso

T2,aniso Furthermore, Eq. 2 suggests that /T2,iso = 1/T2 (550). Hence, 1

1

1

T2,aniso

T2(00)

T2(550)

=

(A2

-

1) 77(55°)

and the results in Table VI show that while 1/ T2 js, increases rapidly as temperature decreases, 1 / T2,aniso increases also, but by a much smaller percentage. If the anisotropic relaxation rate is attributed to the exchange of the bulk water with a very small amount of bound water, we may write4 1

b I

T2,aniso P T2b

(4)

where b is the amount of "bound" water which has preferred orientation. P is the amount of exchangeable water and T2b is the relaxation time of the bound water. At room temperature, P 4A general expression of Eq. 4 may include r (the exchange time), i.e., I / T2.a,m,, = (b/P)/(l1/T2b+ r). However, since the T2 decay at subzero temperatures approaches a simple exponential function, it mat be concluded that r is much shorter than T2b and thus can be omitted.

KASTURI ET AL. Anisotropy in NMR Relaxation Times

377

equals the amount of total water. At lower temperatures, P equals the amount of nonfrozen water which is 25 and 20% of total water at -5 and -1 0°C, respectively. The value of P/ T2,anjso, is given in the last column of Table VI. According to Eq. 4 P

T2,aniso

b T2b

If b is less than the amount of nonfrozen water and, therefore, is not affected by freezing, then P/T2aniso should be constant or increase with decreasing temperatures since (T2b should be shortened as temperature decreases). However, we find experimentally that P/ T2,aniso decreases when the muscle freezes. This implies that b must decrease with decreasing temperature. That is, part of the preferentially oriented water freezes when the muscle sample is frozen. This result suggests that the preferred orientation is a rather long-range effect which extends outside of the hydration (nonfrozen) water layer. Relatively long-range preferred ordering in water molecules has been proposed for some NMR studies of biological systems (46, 56). Woessner and Snowden (56), for example, have observed domains of long-range ordering of polymer molecules of size as large as 15 ,um in which water molecules are preferentially oriented. The data in Table VI show that the isotropic relaxation rate increases by a factor of 15 when the temperature is decreased from 28 to -1 OOC. This increase of relaxation rate may result from the following causes: (a) The nonfrozen water is the hydration water of proteins. Therefore, both the inter- and intramolecular correlation times of these water protons are much longer than that of the total cytoplasmic water (57, 58). (b) The T2 of the cytoplasmic water could be shortened by proton exchange between water and macromolecules (59). Because of an increase in the surface-to-volume ratio at freezing, this proton exchange mechanism would be more effective at subzero temperature. (c) The bulk water freezes into ice at subzero temperatures and its relaxation rate is greatly increased. The proton exchange between the nonfrozen water and ice could also increase the relaxation rate of the nonfrozen water. It seems likely that all of these causes could be operative to some extent and it may well be a combination of these effects which brings about the reduction of the isotropic T2. A detailed investigation of the contribution of these effects may help to gain some insight into the structure and properties of hydration water. Another possible cause of the shortening of T2 at subzero temperatures is that mobile organic molecules may contribute significantly to the proton NMR signal (25, 26). Fung (27) estimated that protons from mobile organic molecules amount to 6-8% of the total nonrigid protons in muscle, and that their T2 at 270C could be of the order of 0.4-2 ms. At temperatures below the freezing point, their contribution to the total NMR signal would be quite significant, and their T2 would exhibit no angular dependence. Consequently, the resultant anisotropy in the T2 data would decrease. On the basis of this argument, the anisotropy should decrease with a decrease in temperature. Berendsen (44) observed a narrow absorption peak in the H20 spectra of collagen hydration, which Dehl and Hoeve (46) showed was not exchangeable with water. This signal could be due to the mobile organic molecules. However, the amount of mobile protons involved is relatively small (roughly 3% of the total macromolecular protons). We believe that Fung's estimate of the concentration of protons from mobile organic molecules is too high. Since the amount of dry solids in a cell is only 378

BIOPHYSICAL JOURNAL VOLUME 30 1980

roughly 20%, Fung's estimate would imply that almost 30% of the dry solid in the muscle would give a mobile proton signal. The available data, in our opinion, do not support Fung's interpretation. We observed very little anisotropy (<1%) in the proton spin-lattice relaxation data in muscle tissue either at 280C or at temperatures below 0°C. Nor did Fung (51) observe any angular dependence of T, for 'H and 2H in frog gastrocnemius muscle at room temperature. The absence of T, anisotropy may indicate that the preferred orientation is associated with the low frequency reorienting motion of the water molecules. It is well known that T, is affected mainly by the motion having a frequency comparable to the Larmor frequency, while T2 is most sensitive to fluctuations in the local fields at low frequency. Our findings would suggest that the anisotropic motion of water molecules must be concentrated in the low frequency range [-c >> (I/cwo) 10-' s]. In conclusion, we have made a quantitative determination of the degree of anisotropy for the bulk of the myoplasmic water as well as for the water that remains unfrozen at temperatures below 0°C. These results demonstrate that a definite but small anisotropy (-5%) exists in the spin-spin relaxation times of water protons in tibialis anterior muscle at room temperature. We find that the anisotropy decreases. to 2.1 % at - 5oC and 1.3% at -1 OOC. These findings suggest that the bulk of the myoplasmic water has a small degree of ordering due to the anisotropic interaction between the water molecules and cellular macromolecules (presumably the actin-myosin filaments). We gratefully acknowledge the assistance of Patsy Seitz during the course of these experiments, and Carolyn Aylward and Mary Comerford for secretarial assistance. This investigation was supported by Robert A. Welch Foundation grant Q-390, The Office of Naval Research contracts ONR N00014-76-C-0100 and N00014-77-C0092, and U.S. Public Health Service contracts GM-20154 and RR-00188. Receivedfor publication 5 April 1979 and in revisedform 6 March 1980.

REFERENCES 1. HAZLEWOOD, C. F. 1973. Physico-chemical state of ions and water in living tissues and model systems. Ann. N. Y. Acad. Sci. 204:631. 2. ABETSEDARSKAYA, L. A., F. G. MIFTAKHUTDINOVA, and V. D. FEDOTOV. 1968. State of water in live tissue (results of investigations by the NMR spin-echo method). Biophysics (Eng. Trans. Biofizika). 13:750. 3. BRATTON, C. B., A. L. HOPKINS, and J. W. WEINBERG. 1965. Nuclear magnetic resonance studies of living muscle. Science (Wash. D.C.). 147:7380. 4. HAZLEWOOD, C. F., B. L. NICHOLS, and N. F. CHAMBERLAIN. 1969. Evidence for the existence of a minimum of two phases of ordered water in skeletal muscle. Nature (Lond.) 222:747. 5. Swwr, T. J., and 0. G. FRITZ. 1969. A proton spin-echo study of the state of water in frog nerves. Biophys. J. 9:54. 6. COPE, F. W. 1969. Nuclear magnetic resonance evidence using D20 for structured water in muscle and brain. Biophys. J. 9:.303. 7. HANSEN, J. R. 1971. Pulsed NMR study of water mobility in muscle and brain tissue. Biochim. Biophys. Acta. 230:482. 8. COOKE, R., and R. WEIN. 1971. The state of water in muscle tissue as determined by proton nuclear magnetic resonance. Biophys. J. 11:1002. 9. FINCH, E. D., J. F. HARMON, and B. H. MULLER. 1971. Pulsed NMR measurements of the diffusion constant of water in muscle. Arch. Biochem. Biophys. 174:299. 10. CHANG, D. C., C. F. HAZLEWOOD, B. L. NICHOLS, and H. E. RORSCHACH. 1972. Spin-echo studies on cellular water. Nature (Lond.) 235:170.

KASTURI ET AL. Anisotropy in NMR Relaxation Times

379

11. HAZLEWOOD, C. F., D. C. CHANG, B. L. NICHOLS, and H. E. RORSCHACH. 1971. Interaction of water molecules with macromolecular structures in cardiac muscle. J. Mol. Cell. Cardiol. 2:51. 12. BELTON, P. S., R. R. JACKSON, and K. J. PACKER. 1972. Pulsed NMR studies in water in striated muscle. I. Transverse nuclear spin-relaxation times and freezing effects. Biochim. Biophys. Acta. 286:16. 13. OUTHRED, R. K., and E. P. GEORGE. 1973. A nuclear magnetic resonance study of hydrated systems using the frequency dependence of the relaxation processes. Biophys. J. 13:83. 14. OUTHRED, R. K., and E. P. GEORGE. 1973. Water and ions in muscles and model systems. Biophys. J. 13:97. 15. HAZLEWOOD, C. F., D. C. CHANG, B. L. NICHOLS, and D. E. WOESSNER. 1974. Nuclear magnetic resonance transverse relaxation times of water protons in skeletal muscle. Biophys. J. 14:583. 16. KNISPEL, R. R., R. T. THOMPSON, and M. M. PINTAR. 1974. Dispersion of proton spin-lattice relaxation in tissues. J. Magn. Resonance. 14:44. 17. NEVILLE, M. C., C. A. PATTERSON, J. L. RAE, and D. E. WOESSNER. 1974. Nuclear magnetic resonance studies and water "ordering" in the crystalline lens. Science (Wash. D.C.). 184:1072. 18. FUNG, B. M., and T. M. MCGAUGHY. 1974. The state of water in muscle as studied by pulsed NMR. Biochim. Biophys. Acta. 343:663. 19. CIVAN, M. M., and M. SHPORER. 1975. Pulsed nuclear magnetic resonance study of 'TO, 2D, and 'H of water in frog striated muscle. Biophys. J. 15:299. 20. FUNG, B. M., D. L. DURHAM, and D. A. WASSIL. 1975. The state of water in biological systems as studied by proton and deuterium relaxation. Biochim. Biophys. Acta. 399:191. 21. DIEGEL, J. G., and M. M. PINTAR. 1975. Origin of the nonexponentiality of the water proton spin relaxation in tissues. Biophys. J. 15:855. 22. BARROILHET, L. E., and P. R. MORAN. 1975. Nuclear magnetic resonance (NMR) relaxation spectroscopy in tissues. Med. Phys. 2:191. 23. RAAPHORST, G. P., J. KRUUV, and M. M. PINTAR. 1975. Nuclear magnetic resonance study of mammalian cell water. Biophys. J. 15:391. 24. CLEVELAND, G. G., D. C. CHANG, C. F. HAZLEWOOD, and H. E. RORSCHACH. 1976. Nuclear magnetic resonance measurement of skeletal muscle. Anisotropy of the diffusion coefficient of the intracelluar water. Biophys. J. 16:1043. 25. FOSTER, K. F., H. A. RESING, and A. N. GARROWAY. 1976. Bounds on "bound water": transverse nuclear magnetic resonance relaxation in barnacle muscle. Science (Wash. D.C.). 194:324. 26. FUNG, B. M. 1977. Carbon- 13 and proton magnetic resonance of mouse muscle. Biophys. J. 19:315. 27. FUNG, B. M. 1977. Proton and deuteron relaxation of muscle water over wide ranges of resonance frequencies. Biophys. J. 18:235. 28. RUSTGI, S. N., H. PEEMOELLER, R. T. THOMPSON, D. W. KYDON, and M. M. PINTAR. 1978. A study of molecular dynamics and freezing point transition in tissues by proton spin-relaxation. Biophys. J. 22:439. 29. WHIPPLE, H. R. 1965. Forms of water in biological systems. Ann. N.Y. Acad. Sci. 125:251. 30. LING, G. N. 1962. A Physical Theory of the Living State. Blaisdell Publishing Company, New York. 680. 31. LING, G. N. 1972. Water and Aqueous Solutions. R. A. Horne, editor. John Wiley & Sons, Inc., New York. 663. 32. LING, G. N. In The Aqueous Cytoplasm. Alec D. Keith, editor. Marcel Dekker, Inc., New York. 230. 33. DICK, D. A. T. 1966. Cell Water. Butterworth Publisher's Inc., Woburn, Mass. 155. 34. PRIVALOV, P. L. 1968. Water and its role in biological systems. Biofizika. 13:163. 35. DRosT-HANSEN, W. 1969. On the structure of water near solid interfaces and the possible existence of long range order. Ind. Eng. Chem. 61:10. 36. DROST-HANSEN, W. 1971. In Chemistry of the Cell Interface. Part B. H. D. Brown, editor. Academic Press., Inc., New York. 1. 37. DROST-HANSEN, W. 1979. Cell-Associated Water. Academic Press, Inc. New York. 440. 38. COPE, F. W. 1973. Supramolecular biology: a solid state physical approach to ion and electron transport. Ann. N.Y. Acad. Sci. 204:416. 39. DAMADIAN, R. 1973. Biological ion exchanger resins. Ann. N.Y. Acad. Sci. 204:211. 40. DAMADIAN, R. 1973. Cation transport in bacteria. CRC Crit. Rev. Microbiol. 2:377. 41. COOKE, R., and I. D. KUNTZ. 1974. The properties of water in biological systems. Ann. Rev. Biophys. Bioeng. 3:95. 42. HAZLEWOOD, C. F. 1979. A view of the significance and current understanding of the physical properties of cell-associated water. In Cell-Associated Water. W. Drost-Hansen, editor. Academic Press, Inc., New York. 165. 43. DAMADIAN, R. 1971. Tumor detection by nuclear magnetic resonance. Science (Wash. D.C.). 171:1151.

380

BIOPHYSICAL JOURNAL VOLUME 30

1980

44. BERENDSEN, H. J. C. 1962. Nuclear magnetic resonance study of collagen hydration. J. Chem. Phys. 36:3297. 45. BERENDSEN, H. J. C., and C. MIGCHELSEN. 1965. Hydration structure of fibrous macromolecules. Ann. N.Y. Acad. Sci. 125:365. 46. DEHL, R. E., and C. A. J. HOEVE. 1969. Broad line NMR study of H20 and D20 in collagen fibers. J. Chem. Phys. 50.3245. 47. FUNG, B. M., and P. TRAUTMANN. 1971. Deuterium NMR and EPR of hydrated collagen fibers in the presence of salts. Biopolymers. 10:391. 48. FUNG, B. M., and M. M. SIEGEL. 1972. The ordering and relaxation times of water adsorbed on collagen fibers. Biochim. Biophys. Acta. 278:185. 49. MIGCHELSEN, C., H. J. C. BERENDSEN, and A. RUPPRECHT. 1968. Hydration of DNA. Comparison of nuclear magnetic resonance results for oriented DNA in the A,B,C forms. J. Mol. Biol. 37:235. 50. DEHL, R. E. 1968. Broad line NMR study of H20 and D20 in oriented rayon fibers. J. Chem. Phys. 48:831. 51. FUNG, B. M. 1975. Orientation of water in striated frog muscle. Science (Wash. D.C.). 190:800. 52. CHAPMAN, G., and K. A. McLAUCHLAN. 1967. Oriented water in the sciatic nerve of rabbit. Nature (Lond.) 215:391. 53. KLEIN, M. P., and D. E. PHELPS. 1969. Evidence against orientation of water in rat phrenic nerve. Nature (Lond.). 224:70. 54. CARR, H. Y., and E. M. PURCELL. 1954. Effects of diffusion on free precession in nuclear magnetic resonance experiments. Phys. Rev. 94:630. 55. MEIBOOM, S., and D. GILL. 1958. Modified spin-echo method for measuring nuclear relaxation times. Rev. Sci. Instrum. 29:688. 56. WOESSNER, D. E., and B. S. SNOWDEN, Jr. 1973. A pulsed NMR study of dynamics and ordering of water molecules in interfacial systems. Ann. N.Y. Acad. Sci. 194:324. 57. CHANG, D. C., and D. E. WOESSNER. 1977. "Bound water" in barnacle muscle as indicated in nuclear magnetic resonance studies. Science (Wash. D.C.). 198:1180. 58. RESING, H. A., K. R. FOSTER, and A. N. GARROWAY. 1977. "Bound water" in barnacle muscle as indicated in nuclear magnetic resonance studies. Science (Wash. D.C.). 198:1181. 59. FUNG, B. M., and T. W. McGAUGHY. 1979. Study of spin-lattice and spin-spin relaxation times of 'H, H, and "O in muscle water. Biophys. J. 28:293.

KASTURI ET AL. Anisotropy in NMR Relaxation Times

381